PROJECT SUMMARY/ABSTRACT Accurate measurement of cognitive function is essential to research on cognitive aging and Alzheimer's disease (AD). Traditional measurement approaches rely on objective (performance based) and subjective (self-report) assessments of cognition made in clinic or interview settings on a single occasion. Unmeasured sources of within-person variability, retrospective reporting biases and the artificial nature of standard testing environments negatively impact test reliability, measurement accuracy, and ecological validity of standard approaches to cognitive assessment. These problems impede the sensitive measurement of cognitive change and delay detection of clinical conditions, such as amnestic mild cognitive impairment (aMCI) and AD. The overall objective of this application is to develop methods to measure cognitive function of older adults in their daily lives in order to improve the reliability, accuracy, and ecological validity of cognitive measurements. Through these methods, this study will better characterize and identify predictors of the short-term variability and change associated with cognitive aging and dementia risk. The central hypothesis is that repeated ambulatory assessments in daily life will improve the reliability, accuracy and sensitivity of both subjective and objective cognitive measures. The rationale for the proposed research is that the development of more precise and ecologically valid instruments will improve the detection of persons at risk for AD and measurement of treatment effects in clinical trials. Guided by preliminary work using these ambulatory assessments in younger samples and using the repeated measurement burst design to predict cognitive decline, the following specific aims are extended: 1) Test the hypotheses that repeated ambulatory assessments will improve the reliability and accuracy of subjective reports of cognitive function and predict incident aMCI and neuroimaging markers of AD, 2) Test that repeated ambulatory assessments of objective cognitive performance will be more reliable and sensitive for measuring change than conventional in-lab assessments, and that ambulatory memory function will predict transitions to aMCI and neuroimaging markers of AD, 3) Test daily stress as a predictor of worse objective and subjective ambulatory cognitive performance and that this effect is mediated by stress- induced worry and alterations in cardiac autonomic function, 4) Test that individuals whose cognitive performance is more affected by daily stress will have increased risk for cognitive decline and aMCI. The approach is innovative in its use of ambulatory methods to measure real-time determinants of cognitive function in daily life. The research is significant because it will improve the precision, reliability, and ecological validity of methods used to monitor changes in cognitive function. The long-term goal of this work is to develop assessments tools that advance the next generation of early intervention and prevention studies to slow or prevent cognitive decline and progression of AD.